Receive Operator Input Correcting Product Information When Precisiion of Identification of Product Is Less Than Reference Value

1. A point-of-sale (POS) register installed in a store comprising: a processing apparatus comprising: at least one memory configured to store one or more instructions; and at least one processor configured to execute the one or more instructions to: acquire an image captured by a camera installed in the store along with the POS register, the image including a product; determine an identification of the product included in the image by inputting the image to an inference device and receiving as output from the inference device an inference result including the identification of the product; compute, based on a history of POS registers, a prior distribution of the product, the prior distribution indicating a purchase frequency of the product; compute a precision of the identification of the product according to a predetermined equation, based on the computed prior distribution, the predetermined equation being:, P = pR pR + ( 1 - p ) ⁢ ( 1 - S ) , Equation ⁢ ( 1 ) where P is the precision, p is the prior distribution, R is recall, and S is specificity; determine that the precision is less than a reference value; in response to determining that the precision is less than the reference value, display on a display device a user interface showing the identification of the product to the operator and a request that the operator confirm or correct the identification of the product; after displaying the user interface, receive via an input device input from the operator correcting the identification of the product; receive input of credit card information to pay for the product based on the corrected identification; and transmit the credit card information to a server of a credit card company to performing payment processing, such that the credit card information is charged for the product in accordance with the corrected identification.

2. The processing apparatus according to claim 1, wherein R and S are fixed values.

3. The processing apparatus according to claim 1, wherein values for R and S are prepared for each of a plurality of environments in advance, and R and S are determined based on an actual environment at a time of computing the precision.

4. The processing apparatus according to claim 1, wherein the at least one processor is further configured to execute the one or more instructions to compute the prior distribution in an actual environment at a time of computing the precision.

5. A processing method performed by a point-of-sale (POS) register installed in a store, the processing method comprising: acquiring an image captured by a camera installed in the store along with the POS register, the image including a product; determining an identification of the product included in the image by inputting the image to an inference device and receiving as output from the inference device an inference result including the identification of the product; computing, based on a history of POS registers, a prior distribution of the product, the prior distribution indicating a purchase frequency of the product; computing a precision of the identification of the product according to a predetermined equation, based on the computed prior distribution, the predetermined equation being:, P = p ⁢ R p ⁢ R + ( 1 - p ) ⁢ ( 1 - S ) , Equation ⁢ ( 1 ) where P is the precision, p is the prior distribution, R is recall, and S is specificity; determining that the precision is less than a reference value; in response to determining that the precision is less than the reference value, displaying on a display device a user interface showing the identification of the product to the operator and a request that the operator confirm or correct the identification of the product; after displaying the user interface, receiving via an input device input from the operator correcting the identification of the product; receiving input of credit card information to pay for the product based on the corrected identification; and transmitting the credit card information to a server of a credit card company to performing payment processing, such that the credit card information is charged for the product in accordance with the corrected identification.

6. A non-transitory storage medium storing a program executable by a point-of-sale (POS) register installed in a store to perform processing comprising: acquiring an image captured by a camera installed in the store along with the POS register, the image including a product; determining an identification of the product included in the image by inputting the image to an inference device and receiving as output from the inference device an inference result including the identification of the product; computing, based on a history of POS registers, a prior distribution of the product, the prior distribution indicating a purchase frequency of the product; computing a precision of the identification of the product according to a predetermined equation, based on the computed prior distribution, the predetermined equation being:, P = p ⁢ R p ⁢ R + ( 1 - p ) ⁢ ( 1 - S ) , Equation ⁢ ( 1 ) where P is the precision, p is the prior distribution, R is recall, and S is specificity; determining that the precision is less than a reference value; in response to determining that the precision is less than the reference value, displaying on a display device a user interface showing the identification of the product to the operator and a request that the operator confirm or correct the identification of the product; after displaying the user interface, receiving via an input device input from the operator correcting the identification of the product; receiving input of credit card information to pay for the product based on the corrected identification; and transmitting the credit card information to a server of a credit card company to performing payment processing, such that the credit card information is charged for the product in accordance with the corrected identification.